The evidence-based United States Public Health Service Guideline on the treatment of tobacco use and dependence (USPHS Guideline) recommends that physicians identify their patients' smoking status, advise those that smoke to quit, assist them in their quit attempts, and follow up on these attempts. Unfortunately, physicians identify only about half of current smokers, advise less than half, and assist and arrange follow-up with a small minority. One approach to this problem is to implement a system that reminds the physician to perform the guideline, and that provides the physician details of the guideline elements and local resources during the patient encounter. A computer mediated clinical decision support system ( ) can actively remind the physician to perform a service for a patient, match characteristics of that patient with a knowledge base containing patient-specific recommendations, provide documentation of this service, and a printed personalized instruction sheet for the patient. However, if a is going to be used by the physician and incorporated into the clinical workflow, the computer hardware and software must be easy to learn and to use, and must be reliable and relevant to the physicians' clinical work. Based on recently completed surveys of physicians and clinic managers, the investigators are developing a prototype for the USPHS Guideline in collaboration with Dr. Richard Shiffman of the Yale Center for Medical Informatics. The development of the is supported by an NIH career development award (NCI 1 K07 CA102585-01) to Dr. Marcy. This proposal is for support of a series of laboratory usability tests, a test of the 's validity using an expert panel, and an initial pilot test in three primary care clinics to permit iterative design-test-redesign cycles that will improve the so that it is more likely to be used by physicians. The effectiveness of the developed through this process would be assessed in a subsequent research proposal using a randomized controlled trial design.